View publication
| Title | Predicting Information Credibility in Time-Sensitive Social Media |
| Authors | Carlos Castillo, Marcelo Mendoza, Bárbara Poblete |
| Publication date | August 2013 |
| Abstract |
Purpose - Twitter is a popular microblogging service which has proven, in recent years, its potential for propagating news and information about developing events. In this work, we focus on the analysis of information credibility on Twitter. The purpose of our research is to establish if an automatic discovery process of relevant and credible news events can be achieved. Design/methodology/approach - We follow a supervised learning approach for the task of automatic classification of credible news events. A first classifier decides if an information cascade corresponds to a newsworthy event. Then a second classifier decides if this cascade can be considered credible or not. We undertake this effort training over a significant amount of labeled data, obtained using crowdsourcing tools. We validate these classifiers under two settings: the first, a sample of automatically detected Twitter "trends" in English, and second, we test how well this model transfers to Twitter topics in Spanish, automatically detected during a natural disaster. Findings - There are measurable differences in the way microblog messages propagate. We show that these differences are related to the newsworthiness and credibility of the information conveyed, and we describe features that are effective for classifying informa- tion automatically as credible or not credible. Originality/value - We first test our approach under normal conditions, and then we extend our findings to a disaster management situation, where many news and rumors arise. Additionally, by analyzing the transfer of our classifiers across languages, we are able to look more deeply into which topic-features are more relevant for credibility as- sessment. To the best of our knowledge, this is the first article that studies the power of prediction of social media for information credibility, considering model transfer into time-sensitive and language-sensitive contexts. |
| Pages | 560-588 |
| Volume | 23 |
| Journal name | Internet Research |
| Publisher | Emerald Group Publishing Limited |
| Reference URL |
|